On Tue, 17 Sep 2024 19:04:39 +0200
ichthyo <p...@ichthyostega.de> wrote:

>On 17.09.24 16:15, Will Godfrey wrote:
>> Not a big issue, but I'm wondering if (in CMakeLists.txt) we should make this
>> compiled by default. I always have it enabled here.
>>
>> Indeed, do we need to make it optional at all? The reporting is already user
>> switchable, and no doubt some people have been confused when it seems to do
>> nothing, not having been compiled in.  
>
>Accessing the system clock always costs some performance; it is an overhead
>comparable to an ordered read from an atomic variable, so it's not negligible.
>Thus I would argue that it is not justified to put such a merely diagnostic
>feature on such a critical code path by default.
>
>Maybe it is in fact not relevant or it goes alongside with existing
>synchronisation overhead. But the point I'd make is, it is hard to prove
>such an added feature really does /not/ matter. Think of different audio
>systems, plugin hosts, different platforms. So without an extended
>investigation, we're just guessing and in such a situation I'd err to the
>safe side and not add anything to a hot path which is not strictly required.
>
>-- Hermann

OK, good point.

I think that leaves two practical options.

The obvious one is to hide the 'Log Load times' line in Settings->Switches when
it hasn't been compiled for - not quite sure how we do that!

Another possibility that might be practical is to use the switch setting to
enable/disable the timing code. That way, the only overhead would be a couple
of bool tests. Possibly with a bit of code duplication that could be reduced to
just one bool.

I actually use this looking for sources of Xruns created by a lot of note-on
events at the same time, or an 'expensive' note on at the same time as (say) an
instrument change on a different part.

P.S.
With Kristian's confirmation, and after I've dealt with Art's list of User Guide
typos and errors, I'll set V2.3.3 rc1.
 
-- 
Will J Godfrey


_______________________________________________
Yoshimi-devel mailing list
Yoshimi-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/yoshimi-devel

Reply via email to